Review on digital solutions for heat pump and refrigeration systems: Project: Digital twins for large-scale heat pump and refrigeration systems Work Package 1 - Deliverable 1.1

The project “Digital twins for large-scale heat pump and refrigeration systems” aims to develop adaptable, modular and reusable models for advanced system monitoring, fault detection and diagnosis and operation optimization. In this context, this report aims to provide a review of the state-of-the-art for digital solutions for heat pump and refrigeration systems. The focus of the review is on numerical models applied for system monitoring, fault detection and diagnosis as well as operation optimization. Three types of model-based approaches are characterized and described, namely white-box or physics-derived models, black-box or data-driven models and grey-box models, which combine the two other model types. It is distinguished that white-box models can provide detailed information about a system, which can be applied to monitor, characterize faults and optimize the operation of heat pump and refrigeration systems. Black-box models can also be used for such applications but unlike white-box models, they lack interpretability. The integration of white-box models with black-box approaches can be used to reduce data requirements compared to white-box models and increase its adaptability to different system configurations and operating conditions. From the reviewed studies, it is noted that further research is required where black-box and grey-box models for fault detection and diagnosis models are studied in operating heat pump and refrigeration systems.